3.1. Test method
In this research, uniaxial compression tests were conducted on limestone specimens containing calcite veins using a WAW-1000KN microcomputer-controlled electro-hydraulic servo universal testing machine. In this way, the strength characteristics of the rock mass are studied, and the state of crack development and the damage pattern of the specimens during the test are observed in combination with the AE and DIC techniques. The installation of all test equipment is shown in Fig. 4 and Fig. 5. When the uniaxial compression test is carried out, the loading rate is controlled by controlling the longitudinal displacement, the preloading rate is controlled as 5mm/min, and the loading rate of the formal uniaxial test is 1mm/min. When the axial load drops to 20% of the maximum force, the specimen is judged to have reached the damage state, and the instrument will automatically end the loading, or the loading can be directly ended by manual control in a suitable state to stop the test. In this experiment, we made the loading system, AE system, strain measurement system, and high-speed camera system to manually turn on and off at the same time to control the start and end of the experiment in order to facilitate the processing of subsequent experimental data and control. In this study, the AE technique and DIC2D technique are combined during the uniaxial compression test to obtain data and information including rock time-axis force curves, AE event numbers, and DIC2D digital analysis images during the whole process of rock damage, which can be used to observe the expansion path of new cracks during the experiment, and to analyze the influence of the original veined calcite on the crack expansion form, damage form and strength characteristics of the specimen.
3.2. Analysis of the influence of calcite veins on the crack extension process and damage mode of specimens
In the uniaxial compression test process, the DIC technology with high-speed camera equipment can more directly show the formation of microcracks, development and macroscopic through crack formation process in the form of images. In this study, the variation of the principal strain at the observed surface of each specimen during the whole experiment was analyzed by using the DIC2D system. The comparison between the principal strain images at different moments shows more intuitively the changes of the principal strain at each point of the specimen surface with time, the location of the stress concentration region in the specimen, the detailed process of crack generation, development and penetration, and the way the original structure of the rock mass affects the damage form of the rock mass and the expansion penetration path of new cracks. As shown in Fig. 6, Fig. 8, Fig. 10, Fig. 12, Fig. 14, ①-⑧ show the principal strain states on the specimen surface at some time points for each specimen. They can outline the main information about the special states of stress concentration, generation of major cracks and final damage in the specimen during the whole experiment. It should be noted that the DIC2D system cannot present the information in the area of 6mm width at the top and bottom of the specimen because two rubber sleeves are used to fix the AE probes at the top and bottom of the specimen during the experiment.
Figure 7. Analysis of the state of the JL1 specimen before and after the test and the main strain during the test.(a) Final state of the specimen (b) Initial state of the specimen (c) Principal strain of the specimen at t = 374s (d) Principal strain of the specimen at t = 489s
Figure 7 shows the photographs of the final and initial states of specimen JL1 and the states of the specimen in the DIC2D principal strain images at t = 374s and t = 489s. Combining ①-⑧ in Fig. 6, it can be found that a microcrack along the longitudinal distribution sprouted in the part A of the specimen during ①-⑤, and this crack produced a through main crack L1 along the longitudinal direction of the specimen from top to bottom (as shown in Fig. 7(c)). When the maximum principal strain of the crack L1 reached about 0.05 mm, the longitudinal splitting damage occurred on the right side of the specimen and the stress was redistributed in the orthogonal plane of the specimen. During ⑥-⑧, two longitudinally distributed cracks L2 and cracks L3, one long and one short, were produced along the right side of the remaining part of the specimen (as shown in Fig. 7 (d)). The crack L2 continues to expand along the longitudinal direction of the specimen as the load is applied, and L2 is nearly through when its maximum principal strain reaches 0.043 mm. Comparing the pre-experimental and post-damage photographs of the specimens in Fig. 8 and the main strain images at t = 374s and t = 489s, The locations of the sprouting position B and the penetration position C of the above main fracture L2 were found to partially overlap with the locations of the primary vein-like calcite B' and C' of the specimen shown in Fig. 7(b).
Therefore, the strain and crack development process during the whole uniaxial compression test of specimen JL1 can be divided into 2 processes, the development and penetration of longitudinal crack L1, and the process of sprouting and expansion of crack L2 and crack L3. The crack extension and penetration processes of the specimens are significantly correlated with the presence of primary calcite veins. The extension direction of the main crack L2, the initiation and the termination of it exist in parts that overlap with the position of the primary calcite veins, and the specimen was damaged in a manner consistent with the form of splitting damage under uniaxial compression test conditions.
Figure 9. Analysis of the state of the JL2 specimen before and after the test and the main strain during the test. (a) Final state of the specimen (b) Initial state of the specimen (c) Principal strain of the specimen at t = 372s (d) Principal strain of the specimen at t = 477s
Figure 9 shows the before and after test photographs of specimen JL2 and the states of the specimen in the main strain images at t = 372s and t = 477s. Combining the principal strain states of specimen JL2 at different times shown in Fig. 8 from ① to ⑧, it can be found that the value of the principal strain near the stress concentration region A shown in Fig. 9(c) increases from state ① 0.0032mm-0.0058mm to state ③ 0.0065–0.0092, and then decreases to state ⑤ from ① to ⑤. 0.0025–0.0065, and the value of the principal strain at area B also increases and then decreases. This indicates that although stress concentration and transverse penetration occurred at regions A and B during ①-⑤, no new cracks were generated in the region of stress concentration as the experiment proceeded, but the region A of stress concentration eventually disappeared because of stress redistribution. By comparing the state of the specimen before the test in Fig. 9(b) and the state in the principal strain image at t = 372s in Fig. 9 (c), it was found that the stress concentration areas at A and B in the DIC2D image correspond to the veined calcite A' and B' in the original specimen, respectively. This indicated that during the uniaxial compression test of JL2, the veined calcite A' and B' within the specimen had some influence on the stress distribution of the rock mass, however, it did not have enough ability to induce new cracks in the specimen here.
After the stress redistribution process at the middle position of the specimen in ①-⑤, with the increase of the axial load, the specimen produced a stress concentration at C shown in Fig. 9 (d) in the process of ⑤-⑥. The stress concentration area was gradually connected from C to D, and a main crack L1 developed from top to bottom, and an obvious stress concentration area appeared at G. Comparing the states of the specimen in Fig. 9 (a), Fig. 9 (b) and Fig. 9 (d), it is found that the penetration path CD of the main crack L1 has overlap with the original veined calcite C'D', and the location of the stress concentration region G has overlap with the right end of the original veined calcite B'.
Subsequently, in the process shown in ⑥-⑧, the specimen developed a new crack at E after the main crack L1 penetrated, and this crack developed from point E to point F, forming the longitudinal penetration of the main crack L2 (shown in Fig. 9 (d)), leading to the final damage of the specimen. Comparing the expansion path of the main crack L2 with the distribution state of the original veined calcite of the specimen, it is found that the expansion path of the crack L2 has a longer path overlap with the original veined calcite E'F' in Fig. 9 (b).
In summary, throughout the damage process of specimen JL2, there is an obvious overlap between the regions with more prominent main strain changes presented in the DIC2D images, the penetration paths of the generated new cracks L1 and L2, and the distribution regions of the original veined calcite of the specimen. This indicates that the development of new cracks in specimen JL2 was significantly influenced by the original veined calcite during the experimental process.
Figure 11. Analysis of the state of the JL4 specimen before and after the test and the main strain during the test. (a) Final state of the specimen (b) Initial state of the specimen (c) Principal strain of the specimen at t = 243s (d) Principal strain of the specimen at t = 443s
Analyzing the principal strain image of specimen JL3 in Fig. 10, it can be found that, similar to specimen JL2, the middle region A of specimen JL3 (shown in Fig. 11 (b)) also produced transverse stress concentration during ①-④, and this transverse stress concentration region also gradually disappeared as the experiment proceeded. Comparing the states of the specimens in Fig. 11 (b) and Fig. 11 (c), it is found that region A has more overlap with the position and direction of the veined calcite A' in the original specimen, which indicates that although the veined calcite did not cause an obvious crack in specimen JL3 at this position, it caused an obvious stress concentration region in the specimen at this position.
In the process of ④-⑧, there are several stress concentration points in the upper right side of the specimen at B where new cracks start to develop, and finally a more obvious and main crack L1 that runs up and down along the axial direction of the specimen develops in area B. Comparing Fig. 11 (b) and Fig. 11 (d), it can be found that the stress concentration region B corresponds to the region B' where the original veined calcite are distributed in the specimen. The penetration path of the main crack L1 overlaps with the veined calcite at C', D' and E'. This indicates that the original veined calcite influenced the crack initiation cracking location and the crack penetration path. In addition, during the penetration of the main crack L1, a penetrating crack L2 was also produced at the position from the middle to the top of the left side of the specimen. comparing Fig. 11 (a), Fig. 11 (b) and Fig. 11 (d), it was found that the crack initiation position of this crack coincided with the starting position of the veined calcite A', and the overall expansion path of the crack L2 coincided exactly with the original veined calcite F '. Comprehensive analysis of the whole damage process of specimen JL3 shows that the specimen went through four main stages. Stress concentration in the middle lateral part, stress redistribution, crack L1 emergence and penetration, and crack L2 emergence and penetration. Throughout the process, the emergence and extension of cracks are more obviously influenced by the primary veined calcite.
Figure 13. Analysis of the state of the JL4 specimen before and after the test and the main strain during the test. (a) Final state of the specimen (b) Initial state of the specimen (c) Principal strain of the specimen at t = 198s (d) Principal strain of the specimen at t = 326s
The image of JL4 in Fig. 12 shows that, similar to specimens JL2 and JL3, the stress concentration and diffusion process occurred in the middle of specimen JL4. Analyzing the image information in Fig. 13, it is found that the region of stress concentration A also coincides with the location of the veined calcite A' in the specimen shown in Fig. 13 (b). Comparing the location of the main crack L1 sprouting and the path of its expansion with the final damage state of the specimen shown in Fig. 13 (a), it can be seen that the location of the crack L1 sprouting coincides with the end point B' of the native veined calcite shown in Fig. 13 (c), and its path of expansion coincides with B'C' and finally passes through D ' across the interior of the veined calcite.
To summarize the crack expansion process of specimen JL4, the stress concentration and the location where the sprouting and expansion of the main crack L1 occurred overlap more with the location of the original veined calcite, so the crack expansion of specimen JL4 was also significantly influenced by the veined calcite.
Figure 15. Analysis of the state of the JL5 specimen before and after the test and the main strain during the test. (a) Final state of the specimen (b) Initial state of the specimen (c) Principal strain of the specimen at t = 71s (d) Principal strain of the specimen at t = 316s
For specimen JL5, as shown in Fig. 14 ①-⑤, the stress concentration in the middle of the specimen is not obvious, but it can also be found that in ①-⑤, and especially in ① shown in Fig. 15(c), the principal strain values in these locations of A-G are larger than those in the surrounding area. As in ①, the maximum value of the principal strain at these locations is about 0.0011 mm, while the maximum value of the principal strain around them is about 0.0002 mm. Observing these locations with large principal strains reveals that they are similar to the locations of the main veined calcite A'-G' that shown in Fig. 15(b). This can be explained by the fact that the stress concentration is more likely to occur at the original veined calcite of the specimen.
Combining the final damage state of the specimen shown in Fig. 15(a) and the state of the specimen shown in the principal strain diagram of the specimen at t = 316 s (shown in Fig. 15(d)), it can be seen that as the experiment proceeds, a stress concentration region is generated at the location H on the upper right side of specimen JL5. Eventually, in this region, in addition to the crack L1 and the branch L2 of L1, which are approximately through the longitudinal direction of the specimen, L4 was generated, while on the left side of the specimen JL5, a crack L3 was also generated from the top. Thus, the specimen produced a total of four relatively obvious main cracks developed along the longitudinal direction of the specimen. Observing the crack initiation locations and penetration paths of the four main cracks, there was no obvious overlap with the veined calcite, and the main cracks were mainly intersecting with the veined calcite.
After analyzing the main strain development condition and crack extension penetration path during the test of the above five specimens, the following conclusions were drawn.
(1) The laterally distributed veined calcites (e.g., the laterally distributed structural faces of specimens JL2, JL3, JL4, and JL5) often result in areas of stress concentration in the rock mass at the location of the veined calcites at the beginning of the loading, but these areas of stress concentration ultimately fail to result in transverse cracking in the specimens there.
(2) When the veined calcites are distributed longitudinally (e.g., the longitudinally distributed veined calcites of specimens JL1, JL2, JL3, and JL4), it often leads to the location of the specimen main crack initiation coinciding with the end points of the veined calcites, and the main crack expansion path is also affected by the orientation of the veined calcites.
The above only analyzes the changes of the main strain and the development of cracks in the orthogonal plane (the plane analyzed by the DIC2D system) of the specimen during the uniaxial compression test. The final damage state of the specimen as a whole is shown in Fig. 16.
Observing the overall damage state of the specimens after the test, it can be found that although the composition of the main body and veined calcite of all five specimens is the same, the crack expansion pattern of each specimen is different when the uniaxial compression test is performed under almost the same experimental conditions. This may be caused by the different distribution states and contents of calcite within the specimens. Therefore, the presence of veined calcite does lead to differences in the crack expansion and penetration paths of the specimens as well as in the overall damage pattern. However, it should be noted that the damage pattern of the specimen is more significantly influenced by the longitudinal veined calcite. The cracks generated by the axial load on the specimen were mainly due to the influence of the longitudinally distributed veined calcites and extended along the veined calcites in more sections, their penetration paths were not significantly influenced by the transversely distributed veined calcites. These cracks pass through the interior of the transverse veined calcite and penetrate along the longitudinal direction of the specimen, and the specimen is eventually damaged as a result of the longitudinal penetration of the cracks produced. The damage pattern of these specimens still conforms to the general pattern of uniaxial damage of rocks in the form of monoclinic shear damage or axial splitting damage.
3.3. Analysis of the effect of veined calcite on the strength of the specimen
The strength curve of conventional rock uniaxial compression test is shown in Fig. 17. The whole damage process of the rock specimen is mainly divided into five stages: (1) oa section: microgap compression and density stage, the curve of this section presents the upper concave shape, and its slope shows a gradually increasing law. (2) ab section: the elastic deformation stage, the line shows the form of nearly straight line, the specimen within the micro gap further closed and compacted, the void is compressed. (3) bc section: the initial expansion stage, the curve gradually changes from the straight form of the ab section to a slightly convex curve, and the volume of the specimen changes from compression to expansion, which is the excess of the specimen from the elastic deformation stage to the destruction stage. Within this stage, cracks within the specimen begin to develop. (4) cd section: destruction stage, the curve of this section compared with the curve of bc section, the rise of the curve further slows down, the curve is more convex, the specimen expansion accelerates in this stage, the deformation grows rapidly. (5) de section: deformation and destruction stage after the peak, the curve of this section shows the form of rapid decline, the specimen is sharply damaged, and the energy value is suddenly released1, 2.
In order to confirm whether the load curves of uniaxial compression tests of rock specimens with natural veined calcite in this study are consistent with the load curves of uniaxial compression tests of conventional rock specimens described above, it is necessary to investigate whether the characteristics of their load curves as a whole and of the different segments of the curves (oa segment, ab segment, bc segment, cd segment, de segment) are consistent with the load curves of conventional rock. In this study, the three AE characteristic parameters of impact number, cumulative impact number and amplitude were analyzed to find the six characteristic points o, a-e of the time-axial load curve of each specimen. The number of impacts, i.e., the number of times that the threshold value of 40 Bb is exceeded and causes the system channel to collect data signals. In this study, it is used to reflect the number of times the specimen reaches a certain intensity of AE activity generated per unit time (1 second is a unit time in this study) during the test, which in turn reflects the frequency of microcrack generation in any one unit time, i.e., the activity of internal microcrack development. The cumulative number of impacts is the total number of impacts generated over a period of time from the start of the test to a certain point, and is used to reflect the number of microcracks generated inside the specimen or the degree of damage to the specimen after a period of time. The amplitude value is the maximum amplitude value of the signal wave generated when each AE event occurs, i.e., the intensity of the AE event. The larger the amplitude, the greater the intensity of the AE event and the greater the energy released when the event occurs, which predicts the larger size and scale of the microcrack within the specimen. (Only the amplitude of the signal exceeding 40 dB is counted in this paper) [16]. Therefore, by combining the load-time curve, impact number-time curve, cumulative impact number-time curve, and amplitude-time curve of each specimen, the six characteristic points a-e of the axial load-time curve of each specimen can be found.
The curves of impact number, cumulative impact number, amplitude, and axial load with time for each specimen are shown in Fig. 18. As shown in Fig. 18(a), the curves and data of AE characteristic values and axial load variation with time for JL1 specimen show that from t = 0s to t = 295s, the axial load-time curve and cumulative impact number-time curve of JL1 specimen both show an up-concave shape, and the slopes of both curves are slowly increasing in this time period, from t = 295s to t = 393s, the slope of the curve tends to be stable, and the curve approximates a straight line. At t = 393s, 483s, and 534s, the axial loads of the rock specimens all showed extreme values, and the amplitude and impact number of the specimens also showed extreme values at the above-mentioned moments. Therefore, by combining the peak points or inflection points of each curve in Fig. 18(a), we can get the time corresponding to these characteristic points of the load time curve a-e of specimen JL1 as t = 295s, t = 393s, t = 483s, t = 534s, and t = 553s, respectively. Using the same method, the times corresponding to the characteristic points such as a-e of the time-axial load curves of the five specimens can finally be obtained as shown in the Table 2.
Figure 18. Curves of the number of impacts, cumulative number of impacts, amplitude, and axial load of each specimen with time. (a) JL1 (b) JL2 (c) JL3 (d) JL4 (e) JL5
Table 2
Time corresponding to different characteristic points in the time-axial load curve for five specimens
|
Number of the specimen
|
The time of each feature point /s
|
|
a
|
b
|
c
|
d
|
e
|
|
JL1
|
295
|
393
|
483
|
534
|
553
|
|
JL2
|
232
|
387
|
438
|
513
|
517
|
|
JL3
|
155
|
276
|
334
|
384
|
458
|
|
JL4
|
121
|
237
|
289
|
341
|
342
|
|
JL5
|
127
|
185
|
285
|
311
|
330
|
It should be noted that the axial load-time curves of specimens JL2 and JL3 do not conform to the characteristics of the axial force-time curves of conventional uniaxial compression tests on rocks for some time intervals. For example, at t = 387s and t = 438s, the curve of JL2 suddenly decreased at those points due to the sudden change of the axial load value, and the curve did not rise steadily in the time period from t = 387s to t = 438s, which did not conform to the characteristics of the axial force-time curve bc section of the conventional rock uniaxial compression test, but the analysis of 0s-232 s, 232s-387s, 387s- 438s, and 438s-513s are still consistent with the characteristics of the axial force-time curve in the oa, ab, bc, and cd segments of the conventional rock triaxial compression test, so, t = 387s and t = 438s are still considered as the time nodes corresponding to feature point b and feature point c. Similarly, the axial load of specimen JL3 at t = 384s is not the maximum value of the whole curve. In the time period from t = 334s to t = 384s, the axial load of the specimen does not gradually increase but gradually decreases, however, the curve characteristics of the specimen within 0s-155s, 155s-276s and 276s-334s are comprehensively analyzed, which are consistent with the characteristics within the oa, ab and bc zones respectively, so that t = 384s is still regarded as the time node corresponding to the characteristic point d. As for specimens JL1, JL4, and JL5, their axial load-time curves are more consistent with the development pattern of axial force-time curves of conventional rock uniaxial compression tests, and the five characteristic points a-e can be clearly found. This indicates that the load-time curves of the above five rock specimens containing natural veined calcite all conform macroscopically to the axial force-time curves of uniaxial compression tests of conventional rocks. By comparing the analysis with the AE eigenvalue curves, all of them were able to identify five characteristic points a-e of their load-time curves. Therefore, the presence of the veined calcite did not lead to very significant changes in the development pattern of the load curves of the specimens and the damage pattern of the specimens: the overall load curves of each specimen still conformed to the rule of slowly rising to the peak point first and then rapidly falling. The whole damage process of each specimen still mainly went through five stages: oa section: microgap compacting stage, ab section: elastic deformation stage, bc section: initial expansion stage, cd section: destruction stage, and de section: post-peak deformation and destruction stage.
However, the final load-time curves obtained from the uniaxial compression tests performed on five specimens with similar dimensions of each specimen, the same composition of the main body of the rock and the veined calcite, and almost the same experimental conditions, still differed significantly in value, as shown in Fig. 19. For example, their peak axial loads were 112.45 kN, 78.34 kN, 64.76 kN, 64.51 kN, and 125.33 kN, respectively. and the load curves of specimens JL2 and JL3 also differed from the curve characteristics of the load curves of conventional rocks in some zones. These phenomena indicate that the presence of veined calcite still has an effect on the strength characteristics such as uniaxial compressive capacity of the rock. These are perhaps caused by the different distribution patterns and contents of calcite.
In summary, the presence of veined calcite did not lead to a very significant change in the uniaxial compression damage pattern of the specimens, the development pattern of axial load with time, but still had an effect on the uniaxial compression strength of the rock mass.